Custom Dashboards in OpenLIT

Customizable self-hosted dashboards for LLM observability

Open Source

Developer Tools

Artificial Intelligence

GitHub

Hi Product Hunt 👋 We’re thrilled to launch the latest version of OpenLIT today! 🚀 OpenLIT is the first truly open-source observability and analytics platform for LLM applications, and now, in our latest release, you get fully customizable dashboards. So many of you asked us for custom dashboards — we went a step further and made them fully importable/exportable as JSON, so you can share and reuse them effortlessly. With our new Dashboard Builder, you can drag & drop charts into folders, apply filters and time ranges, and see exactly what matters to you. A few highlights: 🚀 **Custom Dashboards**: Build any chart or folder you need – no vendor lock-in (bring your own OpenTelemetry SDK) ⚙️ **OpenTelemetry-native**: Works with any LLM/AI SDK (OpenLIT, OpenTelemetry, OpenInference, etc.) – just instrument your app and send data. 🔄 **Import/export**: Move dashboards as JSON for easy sharing and backups 📊 **Real-Time Insights**: Track tokens, latency, cost, and user metrics in real time across chains, agents, and GPUs. 🔒 **Self-hosted & Open Source**: Full data control and privacy – integrates with Grafana, Datadog, or your favorite tools (Coming soon!). We built OpenLIT because monitoring AI apps is hard – from debugging chains of models to keeping costs down. Now everyone can visualize the entire GenAI stack in one place. We’ve included a quick demo video below so you can see how easy it is to create custom dashboards. Try it: ⭐ **GitHub**: https://github.com/openlit/openlit 📒 **Docs**: https://docs.openlit.io 💬 **Slack**: https://join.slack.com/t/openlit... If you love it, give us a ⭐ on GitHub and share feedback! This is our third launch on PH, and we’re grateful for the community’s support. Let’s chat in the comments – we’re here to answer questions and hear your thoughts. Thanks for having us! 😊